Cogent Communications Holdings Inc (CCOI) 2020 Q1 法說會逐字稿

Good morning, and welcome to the Cogent Communications Holdings First Quarter 2020 Earnings Conference Call.

As a reminder, this conference call is being recorded.

It will be available for replay at www.cogentco.com.

I would now like to turn the call over to Mr. Dave Schaeffer, Chairman and Chief Executive Officer of Cogent Communications Holdings.

Please go ahead.

Good morning, and thank you very much.

Welcome to our first quarter 2020 earnings conference call.

I'm Dave Schaeffer, Cogent's Chief Executive Officer.

And with me on this morning's call is Jean-Michel Slagmuylder, our Acting Chief Financial Officer.

We are pleased with our results for the quarter and continue to be optimistic in the underlying strength of our business and our outlook for 2020 and beyond, even with the impact and uncertainty of the COVID-19 pandemic.

Our quarterly gross margin reached a company record high of 60.5%.

It increased sequentially by 20 basis points, an increase year-over-year by 70 basis points.

Our EBITDA margin for the quarter increased 30 basis points from the first quarter of 2019 to 35.8%.

On a constant currency basis, we achieved sequential revenue growth of 0.6%.

And year-over-year, quarterly revenue growth of 5.6%.

A reduction in Universal Service fee excise taxes for this quarter reduced our revenue by approximately $600,000.

On a constant currency basis and adjusting for the impact and a reduction of USF, our quarterly revenue growth would have been 1% on a sequential basis.

Our year-over-year traffic growth was 36% for the quarter, and we achieved sequential traffic growth for the first quarter over the fourth quarter of 2019 of 12%.

During the quarter, we returned $30.6 million to our shareholders through our regular quarterly dividend.

We did not purchase any common stock during the quarter.

At quarter end, we had $34.9 million available in our stock buyback program which our Board has authorized to continue through December 2020.

Our gross leverage ratio decreased in the quarter to 4.78 from 4.86 from the last quarter and our net leverage ratio increased slightly to 2.92 from 2.86.

Our cash held at Cogent Holdings at quarter end was $80.5 million.

This cash is unrestricted and available for dividends and/or buybacks.

Cash held at our operating company level was $294,600,000, and our total combined cash was $375,100,000 at quarter end.

We remain confident in the growth potential and cash-generating capabilities of our business.

As a result, as we indicated in our press release, we have announced yet another sequential increase in our quarterly dividend, increasing that dividend $0.02 per share sequentially from $0.68 -- from $0.66 per share per quarter in the fourth quarter to $0.68 per share per quarter in the first quarter of 2020.

This represents our 31st consecutive sequential increase in our regular dividend.

Now, I'd like to take a moment and just touch on some of the uncertainty related to the global pandemic and COVID-19.

Like many companies, Cogent has been impacted by COVID-19 pandemic effects and the accompanying responses of governments around the world.

Beginning in mid-March, Cogent transitioned its entire workforce worldwide to a work-remote workforce, which continues through to today.

This transition, while difficult, was accomplished with little disruption, and I'd like to thank and compliment the entire Cogent workforce for a seamless transition.

In particular, our IT group stepped up and helped equip our remote workers so they can continue to work effectively.

I'd also like to personally thank our field engineers who are frontline supporters and are out installing customers, maintaining and upgrading our network as we continue to serve customers.

During the first quarter of 2020, the impact of COVID-19 pandemic was fairly limited.

In the last 2 weeks of March, we saw a positive impact in our NetCentric revenue growth, but a slight downtick in our ability to install new corporate customers.

We also saw during these 2 weeks, a material increase in traffic on our network.

However, the ultimate impact of the pandemic on Cogent is unknown at this time, as a significant amount of uncertainty and volatility remains.

While Cogent has moved to a remote work status, we have no assurance that over the long run, this will be sufficient to protect our workforce and key employees.

Moreover, as a result of operations could adversely be impacted as the future of the pandemic and its trajectory and associated governmental restrictions may continue, we may see a downturn in new customer orders, find it difficult to collect from customers who are experiencing financial distress and encounter difficulties assessing -- accessing buildings and locations to install new customers and to perform repairs for existing customers or have difficulty in procuring and/or shipping equipment to our field operations.

Lastly, the global impact of the pandemic may have prolonged economic effects that could impact our business as well as other businesses.

These risks are described in more detail in our 10-Q filed for the quarter and this will be filed shortly after today's call, and these will supplement the risks that are outlined in our year-end 10-K for December 31, 2019.

Throughout this discussion, we'll highlight several operational statistics.

I will review in greater detail certain operational highlights and trends, and Jean-Michel will provide some additional details on our financial performance.

Following our remarks, we'll open the floor for questions and answers.

Now I'd like to ask Jean-Michel to read our safe harbor language.

Thank you, Dave, and good morning, everyone.

This earnings conference call includes forward-looking statements.

These forward-looking statements are based upon our current intent, belief and expectations.

These forward-looking statements and all the statements that may be made on this call that are not historical facts are subject to a number of risks and uncertainties, and actual results may differ materially.

Please refer to our SEC filings for more information on the factors that could cause actual results to differ.

Cogent undertakes no obligation to update or revise our forward-looking statements.

If we use non-GAAP financial measures during this call, you will find the reconciled -- these reconciled to the GAAP measurement in our earnings release, which is posted on our website at www.cogentco.com.

Now I'd like to turn the call back over to Dave.

Hey, thank you, Jean-Michel.

Hopefully, we've had a chance to review our earnings press release.

Our press release includes a number of historical quarterly metrics.

Our corporate business represents 69% of our total revenues.

Our corporate business grew 7.5% from the first quarter of 2019.

Our NetCentric business, which has underperformed compared to historical averages and was flat compared to the first quarter of 2019.

Our corporate business for the quarter grew sequentially 2% -- did 0.2%.

The change in USF tax rate negatively impacted our corporate business, both on a sequential and year-over-year basis.

The rate change is quarterly and we cannot predict future rates set by USAC.

Our NetCentric revenue performance improved from previous quarters and grew by 1%.

The volatility of foreign exchange primarily impacts our NetCentric business.

On a constant currency basis, our NetCentric business increased 1.7% from the first quarter of 2019 and increased 1.4% from the fourth quarter of 2019.

Our long-term EBITDA annual margin expansion guidance remains an improvement to be expected of about 200 basis points per year.

Our multiyear constant currency, long-term growth objectives are 10% per year.

Our revenue and EBITDA guidance are intended to be multiyear goals and are not intended to be specific quarterly guidance.

Now I'd like Jean-Michel to cover some additional financial details for the quarter.

Thanks, Dave.

And again, good morning to everyone.

First of all, I also would like to thank and congratulate the Cogent team for the results of their continued hard work and their efforts during another productive quarter for Cogent.

We analyze our revenues based upon network type, on-net, off-net and noncore and we also analyze our revenues based upon customer type.

We classified all our customers into two types, NetCentric customers and corporate customers.

Our NetCentric customers buy large amounts of bandwidth from us in carrier-neutral data centers.

Our corporate customers buy bandwidth from us in large multi-tenant office buildings.

Revenue from our corporate customers for the quarter grew sequentially by 0.2% to $97 million and grew year-over-year by 7.5%.

We do believe there will be a future impact on our corporate business because of the lower USF rate and corporate customer buying patterns in light of COVID 19.

We had 48,499 corporate customer connections on our network at quarter end, an annual increase of 2.6% over the first quarter of 2019.

Quarterly revenue for -- from our NetCentric customer increased sequentially by 1% to $44 million and was flat year-over-year.

We saw a significant uptick in traffic growth in the last 2 weeks of the quarter, resulting in revenue growth and are continuing to see elevated levels of traffic so far this quarter.

On a constant currency basis, our quarterly revenue from our NetCentric customer increased by 1.4% sequentially and increased year-over-year by 1.7%.

We had 38,740 net-centric customer connections on our network at quarter end, an increase of 9.8% over the first quarter of 2019.

Our NetCentric revenue growth experienced significantly more volatility than our corporate revenues due to the impact of foreign exchange, larger customer size and certain seasonal factors.

Our on-net revenue was $103.5 million for the quarter, a sequential quarterly increase of 0.8% and a year-over-year increase of 6.5%.

Our on-net customer connections increased by 0.8% sequentially and increased by 5.8% year-over-year.

We ended this quarter with 75,163 on-net customer connection on our network in our 2,823 total on-net multi-tenant office and carrier-neutral data center buildings.

Our off-net revenue was $37.3 million for the quarter, a sequential quarterly decrease of 0.4% and a year-over-year increase of 1.3%.

Our off-net customer connection increased sequentially by 0.5% and increased by 5.2% year-over-year.

We ended the quarter serving 11,721 off-net customer connection in our 6,930 off-net buildings.

These buildings are primarily in North America.

Some words about our pricing.

Consistent with historical trends, the average price per megabit of our installed base and our new customer contracts decreased for the quarter.

The average price per megabit for our installed base declined sequentially by 8.4% to $0.53 and declined by 21.8% from the first quarter of 2019.

The average price per megabit on our new customer contract for the quarter declined sequentially by 29% to $0.20 and declined by 49.8% from the first quarter of 2019.

In respect of ARPU -- our on-net ARPU -- sorry, ARPU was flat to the quarter and our off-net ARPU decreased.

Our on-net ARPU, which includes both corporate and NetCentric customers, was $461 for the quarter, the same as last quarter.

Our off-net ARPU, which is comprised of predominantly customer -- corporate customer, was $1,064 for the quarter.

A decrease of 1.4% from the last quarter.

Churn, our on-net unit churn rate was flat for the quarter and our off-net unit churn rate slightly increased.

Our on-net unit churn rate was 1.1% for this quarter, the same rate as last quarter.

Our off-net unit churn rate was 1.2% for this quarter, a slight increase from 1.1% last quarter.

We offered discounts related to the contract term to all of our corporate and NetCentric customers.

We also offer volume commitment discounts to all NetCentric customer.

During the quarter, certain NetCentric customers took advantage of our volume and contract term discounts and entered into long-term contracts for over 3,000 customer connection, increasing their revenue commitment to Cogent by $27.3 million.

Now in respect of our EBITDA and EBITDA as adjusted, our EBITDA and our EBITDA as adjusted are reconciled to our cash flow from operation in all our quarterly earnings press release.

Seasonal factors that typically impact the amount of our SG&A expenses include: The resetting of payroll taxes in the United States at the beginning of each year; annual cost of living or CPI increases; seasonal vacations period; the timing and level of our audit tax -- of our audit and tax services; timing and amount of our gains on our equipment transaction; our annual sales meeting costs and benefit plan annual cost increase.

The seasonal factors typically increase our SG&A expenses in our first quarter from our fourth quarter and result in a sequential increase in our G&A -- SG&A of $3.1 million; fourthly, EBITDA decreased by 4.4% sequentially to $50.4 million.

Our quarterly EBITDA increased year-over-year by $2.8 million or by 6%.

Our quarterly EBITDA margin decreased by 180 basis points sequentially to 35.8%, an increase year-over-year by 30 basis points.

Our EBITDA, as adjusted, include gains related to our equipment transactions.

Our equipment gains were only $39,000 for the quarter, $536,000 for the first quarter of last year, and $251,000 last quarter.

Our quarterly EBITDA as adjusted, decreased by $2.5 million or by 4.8% sequentially to $50.4 million, an increase year-over-year by $2.3 million or by 4.9%.

Our quarterly EBITDA as adjusted margin decreased sequentially by 200 basis points to 35.8% and decreased by 10 basis points year over year.

Our basic and diluted income per share was $0.20 for the quarter compared to $0.16 last quarter and $0.20 for Q1 2019.

Or EUR 135 million notes that was -- that we issued in June 2019 are reported in U.S. dollar and converted at each [one hand] using the euro to U.S. dollar exchange rate.

Realized foreign exchange gain on our Euro notes was $2.9 million this quarter, increasing our basic and diluted income per share this quarter by $0.06 per share.

The unrealized foreign exchange loss on our Euro notes was $4 million for Q4 2019, decreasing our basic and diluted income per share last quarter by $0.09 per share in Q4 2019.

Some words about our foreign currency impact.

Our revenue earned outside the United States is reported in U.S. dollar and was approximately 23% of our total quarterly revenues.

Approximately 17% of our revenue this quarter were based in Europe and about 6% of our revenue were related to our Canadian, Mexican, Asia Pacific and Latin American operation.

Continued volatility in foreign currency exchange rates can materially impact our quarterly revenue result and our overall financial results.

Foreign exchange impact on our reported quarterly sequential revenue was a negative USD 200,000.

And the year-over-year foreign exchange impact on our reported quarterly revenue was a negative $700,000.

Our quarterly revenue growth rate on a constant currency basis was 0.6% sequentially and 5.6% year-over-year.

The variability of foreign exchange primarily impacts our NetCentric customers and revenues.

The average euro to U.S. dollar rate so far for this quarter is $1.09 and the average Canadian dollar exchange rate is $0.71.

Should these average foreign exchange rates remain at the current average levels for the remainder of our second quarter of 2020, we estimate that the foreign exchange conversion impact on our sequential quarterly revenues for our current quarter would be a negative of USD 600,000, and the year-over-year foreign exchange conversion impact on quarterly revenue will be a negative of USD 1 million.

Adding customer concentration, we believe that our revenue and customer base is not highly concentrated.

Up to 25 customers represented less than 6% of our revenues for this quarter.

Adding our capital expenditure, our capital -- our quarterly capital expenditures increased by 30% sequentially due to seasonal factors and decreased by 3.2% year-over-year.

Our capital expenditure were $12.9 million this quarter compared to $13.3 million for Q1, $9.1 million for Q4 2019.

Our capital lease IRU obligations are for long-term dark fiber leases and typically have initial terms of 15 to 20 years or even longer or often -- and often include multiple renewal options after the initial term.

Our capital lease IRU fiber lease obligation totaled $167.9 million at March 31, 2020.

At quarter end, we had IRU contracts with a total of 251 different dark fiber suppliers.

Our quarterly capital lease principal payments under our dark fiber IRU agreements increased by 200% sequentially.

Again, due to seasonal factors, an increase by 103.5% year-over-year.

Our capital principal payments were $6.2 million for the quarter compared to $3 million for Q1 2019 and $2.1 million for Q4 2019.

Our capital lease principal payments, combined with our capital expenditure were $19 million this quarter compared to $12 million last quarter and $16.3 million for the first quarter of 2019.

As of March 31, 2020, our cash and cash equivalents totaled $375.1 million.

For the quarter, all cash decreased by $24.3 million.

We returned $42.5 million of capital to our stakeholders this quarter through our $30.6 million of our regular quarterly dividend payment and $12 million that was spent on semiannual interest payment on our debt.

Our quarterly cash flow from operation decreased by 38.3% sequentially due to an increase in interest payments and an increase in prepaid expenses from annual maintenance and other payments, which typically are made in our first quarter.

Our quarterly cash flow from operations decreased by 0.6% year-over-year, our cash flow from operation was $28.5 million for the quarter compared to $28.6 million for the first quarter of 2019 and $46.1 million for the fourth quarter of 2019.

Regarding our debt and our debt ratio, our total gross debt at par, including our capital lease IRU obligation, was $962.1 million at March 31, 2020 and our net debt was $587.8 million.

Our total debt to trailing last 12 months EBITDA, as adjusted, was 4.78 in March -- at March 31, 2020, and our net debt ratio was 2.92.

Regarding our bad debt and DSO, our bad debt expenses was 0.8% of revenue for the quarter compared to 0.8% of revenue for Q4 2019 and 0.5% in the first quarter of 2019.

Our day sales outstanding or DSO for worldwide accounts receivable were 24 days for the quarter, a one day increase from 20 days last quarter.

While collections are generally in line with historical rates, we did notice some delays in the last 2 weeks of the quarter from corporate customer.

I want to take the opportunity to thank again and recognize our worldwide billing and collection team member for continuing to do a fantastic job in serving our customer and collecting from our customer during these challenging times.

I would now turn the call back over to Dave.

Hey, thanks a lot, Jean-Michel.

Now for a couple of words on the scope and scale of our network.

At quarter end, we had over 961 million square feet of multi-tenant office buildings connected to the Cogent network.

Our network consists of over 36,000 metro fiber miles and over 58,000 intercity route miles of fiber.

The Cogent network remains the most interconnected in the world, and we directly connect to over 7,040 networks.

Of these networks, less than 30 are settlement-free peers.

The remaining networks that we connect to are Cogent customers.

We are currently utilizing approximately 35% of the lit capacity in our network.

We routinely augment capacity on parts of our network as we see increases in traffic to maintain these low utilization rates.

For the quarter, we achieved sequential traffic growth of 12% and year-over-year traffic growth of 36%.

We operate 54 Cogent-controlled data centers with over 606,000 feet of space and those facilities are operating at approximately 33% capacity.

Our sales force turnover in the quarter was 4.7% per month, a rate that was significantly better than our long-term average of turnover in our sales force, of 5.6%.

Our sales force productivity improved in the quarter.

Our quarterly sales rep productivity was 4.5 units installed per full-time equivalent per month.

While this productivity rate remains below our long-term average of 5.1% (sic) [5.1], it was a substantial improvement from the 4.1 units per full-time equivalent rep reported last quarter.

Our sales rep productivity rates are impacted by the decline in our rep tenure as well as the fact that our NetCentric reps are typically selling larger ports.

We ended the quarter with 542 quota-bearing reps.

This was a decrease of 6 from year-end 2019.

We ended the quarter, however, with 522 full-time equivalent reps, these are reps that have gone through their ramp and training period.

And this is a substantial increase from the 502 full-time equivalent reps we had at end of year 2019.

We have modified our hiring process during these difficult times and are now able to onboard reps remotely.

We are supporting all of our employees remotely and continue to monitor rep productivity, utilizing our existing metrics.

In summary, Cogent is a low-cost provider of Internet access and transit services and our value proposition remains simple and compelling.

Our business remains completely focused on the Internet and IP connectivity services as well as data center co-location.

Each of these services are a necessary utility for our customer.

Our multiyear constant currency long-term growth target of approximately 10% and our long-term EBITDA margin expansion rate of approximately 200 basis points should continue for the foreseeable future.

Our Board of Directors approved our 31st consecutive increase in our regular quarterly dividend, increasing our dividend by $0.02 per share per quarter to $0.68 per share per quarter in the first quarter of 2019.

Our consistent dividend increase demonstrates our continued optimism regarding the cash flow and revenue-generating capabilities of our business.

We remain opportunistic as the timing and purchase of our common stock in the open market.

At quarter end, we had $34.9 million remaining under our current buyback program.

We hope everyone remains safe and healthy during these times.

We value our employee safety and are taking all of the necessary precautions to keep our Cogent colleagues safe in these difficult times.

While we believe we are a beneficiary of a stay-at-home model, we are uncertain about the long-term implications for economies around the world with the large number of employees staying at home and the increased rate of unemployment globally.

Our growth and profitability targets remain intact and we are committed to returning an increasing amount of capital to our shareholders on a regular basis.

With that, I'd like to open the floor for questions.

(Operator Instructions) And your first question comes from Michael Funk with Bank of America.

It's [Jon Schrantz] in for Mike.

On just the slowdown in corporate installs, what have you seen thus far in the quarter?

And do you have an expectation on when that kind of returns to a more normal level, obviously, is somewhat reliant on the shutdown being over?

And then how much of an impact on revenue do you think this will have for the year?

Yes.

So first of all, our corporate revenue did not -- grew in the quarter, but not at our historical growth rate.

We grew about 7.5% year-over-year versus a long-term average of about just under 11%.

That slowdown in corporate growth was impacted by 2 things: First of all, the change in USF revenue.

That is purely a revenue metric, not a unit metric.

That rate resets and is designed to fund rural construction in the U.S. and all of the impact falls on our corporate business.

This was the largest sequential change we have seen in years in that USF rate.

On the unit basis, our reps continue to be productive, and we continue to sell at historical rates.

We did see in the last few weeks of March, the inability to get into certain buildings, particularly in markets such as New York, that were particularly hard hit by COVID-19.

As buildings were trying to protect other tenants' employees, they would not allow construction workers to come in and turn up incremental customers.

We also saw a significant uptick in customers looking for more bandwidth at their corporate locations, which at first blush, sounds counterintuitive, because as the employees are working from home, you would say, why would a company need more bandwidth at a corporate office when there are no employees there.

And in fact, the work from home phenomenon for most corporate customers increases the amount of bandwidth they need in their office, as those customers' ad hoc VPNs are being concentrated at that location.

And in order to make sure that employees can get adequate access to information and not experience bottlenecks, many of our customers have looked to increase those connections from 100 megabit, which is our most prevalent corporate product, to a gigabit, which has become our most prevalent new sale, and we have had to upgrade a number of customers.

As stay-at-home orders are being lifted around the country, our field technicians are continuing to be able to get access to buildings and we expect kind of a rebound to more traditional levels of corporate growth.

And your next question comes from Philip Cusick with JPMorgan.

Can you talk about some of the last 5 weeks of the business since the quarter closed?

You mentioned that installs were tougher in March, but what about churn since then?

Are you still relatively insulated from the SMB weakness?

Yes.

Hey, thanks for the question, Phil.

So on the NetCentric portion of our business, we are providing an input to the service our customers are selling.

And we've seen no increase in churn and we've seen the accelerated rate of traffic growth continuing, albeit the rate of increase in that growth has leveled off at this point.

On the corporate side, we continue to monitor collections diligently.

We are concerned about business failures, but we have an added advantage in that our customer base tends to be more resilient in difficult economic times because they have been vetted by the building owners.

If you look at the roster of buildings that Cogent connects to, by market, they tend to represent the most expensive real estate in that market, Class A skyscrapers, and the tenants that can typically pass the credit checks of those landlords tend to be fairly stable in difficult economic times.

Now we monitor our collections daily.

We saw our rate of collections throughout March and April continue at slightly above the previous year's rate of collections.

Now many of our customers don't have employees in to process bills and make payments.

So we saw a 1-day increase in DSOs last quarter, we have not seen any significant increase in DSOs as of yet in this quarter.

And our bad debt remained at about 0.8% consistent with historical rates.

For the most part, we expect our churn rates and customer failures to remain low.

But we are monitoring that.

And we are going to see material negative GDP growth and material unemployment, and it would be naive to think that none of that could impact any of our customers.

But I think we are in a much better position than virtually any other wireline telecom services company.

And then on the NetCentric side, you said that you saw an acceleration of traffic growth that's starting to moderate off.

Are you seeing that from mostly your biggest customers?

Or is there a nice broadening of the customer base accelerating?

It's actually broadened quite a bit from where it was, say, even 6 months ago.

We're seeing a lot of alternate, over-the-top video streaming services, and that has broadened the base quite a bit.

We've also seen a number of our access networks around the world continue to need more bandwidth for their end users, and that also broadens the base which I think helps us on the effective rate of price decline per megabit.

As Jean-Michel commented on the call, we remain relatively unconcentrated in our customer base.

And I think on the NetCentric side, we're continuing to benefit from a broadening of the growth in traffic from a wider number of customers.

And your next question comes from Sam Badri with Crédit Suisse.

Dave, this is actually George on for Sami.

So I guess my question is, over the past few quarters, you've mentioned that [bolder] sales reps would improve unit productivity.

We also know that larger average port sizes are negatively impacting productivity.

So I guess how should we balance those out for the remainder of 2020?

So a few things.

One, our total number of reps declined slightly in the quarter by 6, yet our full-time equivalents increased by 20, meaning we got a pickup in tenure, at least in that band that have been here more than 3 months, and were trained.

Now quite honestly, we were unable to onboard reps in the last 3 weeks of March.

It took us about 3 weeks to modify our onboarding so we could still continue to hire without the reps coming into a physical office for training.

We actually have the largest backlog of reps to start that have accepted offers in our history.

We have an online, video-based training system where our regional learning managers are conducting classes, and we're trying to keep those video classes to about 20 individuals, whereas the physical classes tended to be larger and at more like 40, as we experiment with what is optimal to bring those reps on.

We expect our head count growth to continue and to hit our targets of a 7% to 10% aggregate growth for the year.

We also expect our average rep tenure over the year to continue to improve.

We also have been monitoring our ability to manage out underperforming reps in a remote environment.

We have been concerned about the fact that managers don't see physically every rep in person every day.

Every manager is required to do 2 video conferences with their entire team each day.

And then one-on-ones with individual team members.

But that has us concerned that maybe the underperformers aren't getting the attention they need.

So we put some additional efforts in place to help there.

All in all, we expect rep productivity to continue to increase the movement from 4.1 to 4.5, was a material improvement, but still slightly below the 5.1 long-term average.

We expect that to continue to improve through the quarter and we expect to see our full-time equivalent number continue to grow.

Now for NetCentric reps, they are selling larger ports.

That is a slight offset.

The USF drag is an offset on ARPU, but all in all, we feel pretty good about our sales force improvements.

Got it.

That's helpful.

And then another one on, I guess, capital allocation.

So given the way that your business has been scaling, are you considering increasing the rate of dividend increases from $0.02 to maybe $0.03?

And then in previous quarters, you've also mentioned that you've always used the entire stock buyback authorization.

So I guess, how should we think about that going forward as well?

So we had a board meeting yesterday and our Board, every quarter, debates the reallocation of capital and the mechanism to do that.

We are concerned about macro volatility but confident in our business.

Today, our current return policy feels appropriate.

As we project out in the future, it is clear that we will need to return more capital.

What that mechanism is, buybacks versus dividend, is highly dependent on market volatility.

While we saw a little bit of volatility early in the crisis, much of that volatility again has dampened down.

So this is a long-term analysis, not a specific quarterly decision and it's also one the Board takes very seriously as they know once they increase that pacing, that we will be expected to maintain that increased rate of dividend growth.

With 31 consecutive quarters of dividend growth and the commitment to do that in this environment, I think Cogent remains literally one of the handful of companies globally that is committed to this type of dividend growth policy.

So it's just going to be evaluated each quarter.

I don't have a specific answer, but we also have leverage targets, and we're going to remain within those leverage boundaries.

Your next question comes from Frank Louthan with Raymond James.

Sorry about that.

I'll learn to use the mute button a little better next time.

A couple of questions.

Looking at the trends in network traffic.

How much of that traffic that you've seen is expected to be permanent?

And kind of given the abrupt traffic increase, why aren't we seeing a bigger bump in the revenue?

And how should we think about that going forward?

All right, I'm going to take those in reverse order Frank.

First of all, the step-up in the rate of traffic growth occurred the last 2 weeks of the quarter.

So it's not going to have a material impact.

And remember, it only impacts 31% of our revenue.

So 31% of our revenue saw an uptick for 2 weeks out of 13 in the quarter.

So a fairly minor impact.

It should have a much more material impact as that carries through in Q2 and beyond.

Now to the part around permanency of this uptick in traffic.

We support a number of key applications, video conferencing, audio conferencing, and all of that traffic has materially increased as we've gone to a more work from home environment.

People will eventually return to their offices and there will be a reduction, maybe not a complete reversion to where we were before, but a reduction in that type of traffic.

That traffic is de minimis compared to streaming video traffic, which is the primary driver of unit volume growth.

We see this broad mix of OTT business models, accelerating their displacement of linear television.

That is a permanent trend, not a temporary trend.

What is temporary is people may be watching more minutes a day of video, but what is permanent is the migration from linear to over the top.

And while we saw a material spike up in the rate of acceleration, that rate of acceleration has returned to a more normalized rate of acceleration, but we're at a -- off of a higher base.

And we do expect our rate of traffic growth for the full year 2020 to be above that of 2019.

All right.

Great.

And then you -- I believe you were offering your corporate customers maybe $200 or something to that effect, of additional revenue per month.

You were offering them some unlimited bandwidth or a significant increase in bandwidth, something to that effect if they took it for the rest of the contract.

What percentage of your corporate customers took that?

And has that continued?

Have you continued to see that uptake into Q2?

And do you think that's peaked?

Or should we expect to see more corporate customers take advantage of that offer?

Sure.

So we saw a number of customers, starting at mid-month when we rolled out the promotion, which continues, taking advantage of that upgrade, the ability to take their current connection, increase it from 100 megs to 1,000 megs for an increase in monthly recurring revenue of $200.

They do not have to extend their contract terms.

So it's for whatever the remaining term is.

About 2% of our base has so far taken that.

About 11% of our base, prior to mid-March, was at already a 1 gigabit speed.

About 2% of the base was somewhere between 100 meg and 1 gig, some fractional gig service, and 87% of the base was at 100 meg.

We've seen about 2% shift.

So the 85 has got -- or 87% has gone to 85% and we're continuing to roll that out.

I think as companies continue to understand that some portion of their workforce will be remote for a long period of time due to the need for social distancing within the office, even as they return to work, we expect this rate of conversion to continue.

And it should be helpful to our corporate on-net ARPUs, which, in fact, were flat compared to the historical rate of a slight decline, due to contract lengthening, and it's a result of this incremental revenue from the $200 uptick as they upgrade to gigabit speeds.

All right, great.

I guess one last question.

Would you consider changing some of your reporting to report revenue and so forth on NetCentric versus corporate?

I mean I think that's how most people think about the business.

We certainly talk about it that way, and on-net and off-net made a little more sense a decade ago, but how should we think about that?

And any thoughts about changing how you report and describe the business publicly?

Sure, Frank.

So we actually classify each dollar of revenue 4 different ways.

On-net versus off-net, which is critical for margin contribution; by product, which is critical for things such as USF impact; we do it by geography, which is important for FX; and by customer type, which is important for addressable market.

We are careful not to add so much detail that we obfuscate things for investors.

We do give that breakdown in our Q, where you can see the growth by customer type.

We've just been reluctant to add more metrics and confuse investors.

We, in the future, may evaluate looking at all of the ways in which we report, but for 16 years as a public company, we've really tried to be exceedingly consistent.

And your next question comes from Walter Piecyk with LightShed.

Dave, is excise taxes, is that where your USF is booked, the $3.7 million?

That is correct.

Excise -- that is our only excise tax.

Got it.

So I mean, if I added, I mean it looks like you were down $600,000.

If I put you up $200,000.

So delta of it, $800,000, your corporate still only did 100 basis points sequential.

So it's obviously not USF.

That's the only headwind to corporate.

Can you talk a little bit more about the things that took a 2.5% -- or excuse me, 2.8% sequential growth business down to 1%?

Yes.

I think it was 2 additional factors, Walt, and I think you're absolutely correct in the way in which you did your arithmetic.

But we did experience a inability to install new customers in the last few weeks of March.

As we eventually were able to get our contractors and our field people able to work, we had to get certifications from the FCC that these were essential services, that these people could travel and go into buildings that had stay-at-home orders.

So we did miss about 3 weeks of full install productivity.

That's part of it.

And then...

Three weeks?

Where was there shelter in place for 3 weeks?

New York, I believe...

I don't think so.

I don't think so.

I think it was more like 2 weeks.

I thought it was like the 12th of -- so we went on...

Definitely not the 12th.

We went on the 13th as a company, globally, to work from home.

So we lost 18 days.

I mean, that's a little bit of it.

And I think it was a softer quarter, was all of those things.

And if you look at the corporate business, it tends to be fairly steady, but there is some volatility.

And this was not the best corporate quarter.

Your next question comes from Colby Synesael with Cowen.

Great.

Thank you.

Actually, just a few quick ones.

First off, you mentioned that you expect traffic, I assume you're speaking specifically to NetCentric, to be up in '20 versus '19.

Can you tell us what the bogey is?

What was it in '19 that we should be kind of comparing to?

Second, on the 100 meg to 1 gig, you said that that's a $200 million -- $200, excuse me, per month incremental rev opportunity for you guys.

I assume that, that's something you guys could do on the back end, and therefore, does not require installs, or for people to actually go to those buildings.

Therefore, that should be happening fairly quickly as those orders come through.

Just wanted to confirm that.

And then the last question, what percentage of your incremental revenue growth each quarter typically comes from -- and I'm talking about corporate, comes from new customers versus current customers?

And what are you seeing?

I mean it sounds like just in response to Walt and others, you're expecting that the installs to improve and that we can get back to business as usual, if you will, by the second quarter?

I'm just curious if I'm hearing that correctly.

Yes.

All right.

Let me try to take all of those.

Let's start with the traffic growth one.

It was about 34% last year.

It will be better than that.

Our long-term average is around 47%.

I'm not sure we'll quite be there, but we will be up over 34% year-over-year and that is driven almost entirely by NetCentric.

Secondly, the conversion to a gigabit interface typically does not require us to do anything, but it does, in many cases, require the customer to remove a media converter at their premise.

So we deliver a fiber connection to the customer's premise.

And we rate limit that to 100 megs.

That's done in our equipment in the basement.

Many customers historically could not accept a fiber handoff.

So there is a small device, it's about the size of a pack of cigarettes, that is installed in the customer suite.

It is part of our install charge, it generally costs us about $250, and it is a media converter, taking that fiber signal and converting it to copper.

That device needs to be removed, so you literally pull the copper jumper from that device to the customer's router or switch out, and that device, and plug the fiber connection in.

We walk the customer through that.

In most cases, the customer can do it themselves.

In some cases, we do dispatch a field person to do that, but it does take someone to physically be at the site.

There is a very small number of corporate customers who were taking 100 meg connections on a fiber handoff.

In that case, we don't have to do anything.

There is zero physical work, but that is not the most common factor.

And then finally, to your question about corporate growth rebounding, part of the downturn, too, and I probably should have called this out on my answer to Walt, is that we did see also a slowdown in off-net corporate installs.

You saw our corporate growth was slightly lower than -- our corporate off-net growth, virtually all off-net is corporate, was lower than what it's historically been.

And our roughly 90 vendors have had their own challenges with their workforces and being able to turn up those customers.

We're seeing most of those issues now resolved and we're returning to more normalized off-net installation windows.

So would you expect then just to kind of hit the -- with the -- just be a more specific here.

But I mean are you -- are the install issues, are they abated at this point?

Or should we still expect some type of moderate headwinds or pressure in the second quarter?

I think they're pretty well abated at this point and we should expect the second quarter to look more like a normalized quarter on corporate growth and unit installs.

And your next question comes from Nick Del Deo with MoffettNathanson.

First, you noted the NetCentric base has broadened quite a bit, but it looks like the pricing for new NetCentric contracts is quite a bit lower than you've recently seen.

I think Jean-Michel said it was $0.20 in the quarter.

Can you help square those observations and what it might mean for the pace of NetCentric revenue growth?

Sure.

So that pricing number that we quote and have always quoted, is literally on a per contract signed basis, and there is a disconnect between the volume-weighted versus the nominal price.

So you can sign a lot of contracts at a low price, but your growth can come from customers who are under contract at higher prices.

In fact, that's the phenomenon we saw in the quarter.

So while we had a greater than normal sequential rate of growth or decline at 49% versus what's normally around 15%, the year-over-year was pretty much in line.

We signed some contracts at lower prices, but most of the growth was actually coming from customers with existing contracts at a higher price.

So the delta between the contract price and the realized price narrowed.

Okay.

Got it.

That makes sense.

And then one more on the NetCentric side.

You oftentimes structure those contracts so they have pricing breakpoints to incent customers to get, give you higher traffic volumes.

I think in the past, you've used the analogy of like a rabbit at the dog track.

Has the spike in traffic growth pushed any customers much closer to those thresholds, faster than expected, such that there might be some effect discernable on the numbers?

Or is that not something that we should be worried about?

Well, it's something we monitor.

And I guess, it was probably about 1.5 years ago, we had at least 1 or 2 large customers that hit some pretty meaningful thresholds and it impacted us.

So we've kind of tried to ladder those out more judiciously, and we've tried to make those deltas smaller.

A NetCentric customer expects 3 things.

One, a lower price as time goes on.

That's independent of volume.

Two, they expect that if their volume growths are substantially greater than that of the market, they expect a bigger rate of decline.

And then third, just aggregate volume.

So not only the rate, but the aggregate number gets them to a lower price.

The good news is, we have seen a lot of what we would classify more as mid-tier NetCentric customers growing at accelerating rates, which has actually helped us on this volume-weighted effective price per megabit, as I touched on in the first part of your question.

Your next question comes from Brandon Nispel with KeyBanc.

Dave, I'm curious about your view.

Just from a macro perspective, with everybody working from home, I guess, do you think enterprises and corporate customers change their leasing patterns for multi-tenant office space, longer term?

Second, I think I missed the churn metrics, if you could provide those, that would be great.

And then third, can you help us think about whether or not the slowdown in installs impacts capital spending for the year and maybe where you expect CapEx to come in at for 2020?

Sure.

I'll probably take those in reverse order.

Let's start with CapEx.

And I think it will be pretty much in line with last year, down slightly from the year before, but we are expanding our corporate footprint at a slower rate.

That's why we were down a little bit last year and expect that to continue this year.

Secondly, on the churn rates, the on-net churn rate is about 1.1% per month on a unit basis, off-net about 1.2%.

I do think the off-net is probably a little more vulnerable to some of this reordering of workspace size than on-net.

But it's not a material difference.

And then the final question, which is more of a REIT-oriented question, which is, are people going to go back to their offices when they need to, or want to, or can?

I think there's kind of 2 different schools of thought.

One is, people permanently have a larger percentage of their workforce population work remotely.

Secondly, for those that go back, they'll actually need more space, because of the need for social distancing, and this may be the end of the open floor plan and a return to more physical offices, which tend to be a higher amount of square foot per capita of employee.

I don't think we have enough information at this point to know which way that's going to go.

What typically happens in a down market is rents decline materially, the Class A office space tends to remain full, but at lower rates.

And really the surge in vacancy tends to be in the more marginal office space, in the Class C and Class B buildings.

So I'm not overly concerned that we'll see big office towers go dark.

We did not see that in the Great Recession, in 2008, 2009, very different factors causing it, but that was actually an economic downturn that hit the financial industry the hardest, and we still saw financial services companies maintain their office space.

This economic downturn is hitting Main Street much more, and that tends to be kind of retail space and strip malls and low-rise office parks, and I just don't think we're going to have a big issue with skyscrapers not having businesses in them.

And your next question comes from Tim Horan with Oppenheimer.

Three questions.

One, did you experience any congestion on your own network or handing off traffic to peers on the IP transit business?

And there's a lot of articles in Europe that people are asked to throttle back.

And I guess, could your traffic growth have been a little faster?

Secondly, I agree that the Class A office space is going to stay full.

It's just that I think there's going to be a lot of churn of the tenants in there.

And I guess the question, if that's kind of a short term negative, but can you talk about your flow share when new tenants kind of come in versus your existing market share?

And then lastly, a million moving parts.

Can you maybe give us a little more color for this year on kind of what the revenue trends look like?

I know you don't give guidance, but I mean, this 5% roughly, is this a good way to think about the year and all the puts and takes.

Yes, sure, Tim.

And I'm going to take those in reverse order.

I'll start by saying our multiyear guidance is 10%.

This year feels like it's going to be much closer to that guidance than the 5.5% that we did this quarter on a year-over-year basis.

We do have, I think, some tailwinds to our business that became obvious in the latter part of Q1 and will be meaningful as they materialize for the full quarter in Q2 and continue throughout the year.

So we feel very good about improving revenue growth.

To the question about tenant turnover, that actually would be a tailwind to us because we have about 25% penetration in the buildings that we serve, we -- the average building has approximately 51 unique tenants on the Cogent network.

We have sold 23 connections to just under 15 of those tenants.

And when a tenant moves into a building, they have to make a bandwidth purchase decision.

When we get in front of someone who has to make a decision, we win 50% of those opportunities due to the speed of our install, the higher reliability, the diversity and the greater throughput.

So we expect to do better when there is higher turnover in those buildings.

So overall, I think we'll be in good shape as we may see some greater rate of new leases in buildings.

Now to the congestion question, and there were really 2 questions in that question.

The first one was, a number of our customers were asked to voluntarily reduce resolution to lower strain on last-mile networks in Europe.

That did reduce bandwidth from those customers compared to what it would have been if they continued to stream in higher resolution.

In aggregate, I'm not sure that was a very material impact, but it could have been very meaningful for those on the access networks that were being congested.

In terms of our interconnections, the vast majority of our traffic goes customer to customer.

In fact, about 73% of our traffic now goes from a Cogent access network to a Cogent content producer or conversely, only about 27% traffic goes through peers.

Many of our access network customers, those 7,000 access networks were scrambling to upgrade.

Some of them were in very exotic markets and were dependent on subsea capacity being upgraded.

So there was a bit of a lag there.

And then on our peers, many of our peers tended to be larger, more bureaucratic entities and did not react as quickly as would have been optimal.

The good news is, we had plenty of headroom with everyone, with the exception of Deutsche Telekom and our ongoing litigation.

All of the others were fine, but I would say that some of the incumbent operators moved not as quickly as would be ideal, but we did not and still do not experience any congestion with any access network, other than Deutsche Telecom.

And your next question comes from Bora Lee with RBC Capital Markets.

Two, if I could.

So the first one is the principal payments of finance lease obligations was about $6.2 million in the quarter versus $3 million the prior year.

Can you just give some color on the reason for the year-over-year increase?

And then...

Go ahead.

And then secondly, there was some optimism about the impact streaming app launches could have on network traffic.

Is there any way to unpack how network traffic was trending prior to COVID in 1Q, and how that looked relative to 4Q?

Yes, sure.

So let me take the IRU principal one.

That's actually pretty easy.

We actually had a pretty material uptick in our footprint.

We added almost 400 miles of metro fiber and a few, I don't know, almost 400 miles also of long-haul fiber, much of that was prepaid.

We did some major expansions in Turkey and Brazil, a little bit here domestically, but our network actually grew substantially to over 94,000 miles.

And most of that was prepaid.

So that was the biggest piece of it.

A little bit of it is timing, but most of it's just prepaid IRUs.

And since there's no interest, you're prepaying it, it all goes to principal.

In terms of the traffic growth, Q1 looked like a very normal quarter for the first, probably 10 or 11 weeks of the quarter.

And we really saw the abnormal spike in traffic only those last couple of weeks, and that made the quarter grow sequentially a little bit faster, at 12% versus probably would have been a 7% or 8% sequential grower.

And on the year-over-year, it wasn't quite as meaningful because it was just a couple of weeks, but we do anticipate seeing a full quarter's impact in Q2.

Got it.

And just one last one.

The new corporate install requests and upgrades that weren't able to be installed due to building restrictions.

Do those tend to remain in the pipeline?

Or do they just fall out?

No, they tend to remain in the pipeline.

In fact, customers understand that.

And particularly for off-net ones, we have a fair amount of cushion in our SLA built-in.

But I think most customers tend to be pretty reasonable.

Our installs tend to be faster than industry averages.

And I think people also are realistic when they see, the streets are empty, they understand it could be difficult or they may be under an order from their landlord not to allow people into a building, and they know if they can't get in as a tenant, it's going to be difficult for contractors to get in and hook their service up.

And your last question comes from Mike McCormack with Guggenheim Partners.

Just a quick question.

Just thinking about the Verizon's multi-edge compute initiative and then some of the commentary from the tower companies just sort of getting more compute towards the edge, the tower if you will, tower footprint, perhaps.

What does that mean for your business long term?

I'm a little more skeptical of how impactful that will be.

And the reason is, the space at the base of a tower is very precious and expensive.

Also, the ability to utilize that compute and storage effectively is lower because you're doing it over such a small footprint.

It's kind of an extension of the CDN architecture, and CDNs that have too many endpoints tend to be less efficient than those that have a more moderate number of endpoints for compute, and you're trying to balance the use of that computer, the cost of the storage and power against how frequently the content that is being processed and requested, is accessed.

There is a place for edge computing, but it's not material.

Most traffic will leave the cell site and eventually go to a mobile switching office and because it's IP, to the Internet, as opposed to being fully handled at that site.

I understand carriers are looking for ways to reduce latency and improve load on their backhaul networks.

This will have a marginal impact, but not a material impact.

And then finally, what's most important is, while mobile users represent 80% of the user base globally of the Internet, they still only represent less than 3% of global Internet traffic.

So the vast majority of traffic gets terminated on a wireline connection.

So I don't think this is going to have a meaningful impact on our business.

And I'm showing no further questions at this time.

I would now like to turn the conference back over to the speakers for any closing remarks.

So first of all, I want to thank everyone, and I hope everyone stays safe in this environment.

I actually have one more topic I want to touch on.

I actually thought I would get a question about this.

And this is about liquidity.

We do have $189 million of unsecured debt that matures in a year.

That debt is callable at par.

We are considering calling that debt and refinancing it.

Earlier in the -- actually in the quarter, we were concerned about our access to the debt markets.

Cogent did apply for a PPP loan as a wireline carrier classified by the SBA, we met the criteria as a small business, with less than 1,500 employees.

We did receive that loan, and based on the criteria that was in place at the time that we drew the loan, we were very comfortable with the certifications we made.

On April 23, the SBA modified those criteria.

And quite honestly, the fact pattern in the market had changed.

Based on those criteria, we concluded that it would be prudent for Cogent to return that loan.

So we have, in fact, returned the $10 million that we had drawn under that loan and now plan to look to the public debt markets.

As one of the world's largest Internet providers, we have a service that people are dependent on.

We need to maintain the liquidity that is necessary to run our business and we have not furloughed any employees and, in fact, continue to hire employees.

But we felt that using the PPP in light of the newer market data we have and the additional criteria that the SBA placed, was inappropriate.

So again, I want to thank everyone.

I want everyone to stay well, and we'll talk soon.

Take care.

Bye-bye.

Ladies and gentlemen, this concludes today's conference call.

Thank you for participating.

You may now disconnect.