Cogent Communications Holdings, Inc. (CCOI) Q3 2015 Earnings Report, Transcript and Summary

Welcome to the Cogent Communications Holdings Third Quarter 2015 Earnings Conference Call. [Operator Instructions]. I would now like to turn the call over to Mr. Dave Schaeffer, Chairman and Chief Executive Officer of Cogent Communications Holdings.

Good morning and thank you. Welcome to our third quarter 2015 earnings conference call. I'm Dave Schaeffer, Cogent's Chief Executive Officer. With me on the call this morning is Tad Weed our Chief Financial Officer. We're very pleased with the results for the quarter and are optimistic in the strength of our business and the outlook for the remainder of 2015 as well as for full year 2016. During the quarter, we experienced accelerated year-over-year and sequential constant currency revenue growth, accelerated traffic growth, EBITDA margin expansion, increased cash flow and an increase in our sales force rep productivity. During the quarter, we returned a total of $27.5 million to our shareholders through a combination of our dividend and stock buybacks. During the quarter, we purchased 431,000 shares of our common stock in the open market for a cost of $12.2 million at an average price of $28.20 per share. At the quarter end, we had a total of $47.8 million available under our stock buyback program which continues through December 2016. We continue to remain confident in the growth potential and cash flow generating capabilities of our business. As a result and as indicated in our press release, we announced yet another increase to our regular quarterly dividend from $0.34 a share per quarter to $0.35 per share per quarter, our 13th consecutive quarterly increase sequentially in our regular dividend. Since we purchased a total of $12.2 million of our common stock in the third quarter, under our return of capital program which was greater than our minimum commitment of $12 million, under this program we will not be making a special dividend payment in the fourth quarter. Our return of capital program was scheduled to continue until our net debt to EBITDA as adjusted was within our targeted range and a ratio of anywhere between 2.5 and 3.5 to 1. This ratio increased to 2.98 as of September 30, 2015, from the 2.77 at the end of the second quarter. As noted in prior earnings call, our return of capital program is subject to changes as conditions may warrant. The indentures governing our notes, both senior and our secured notes, limit our ability to return cash to our stockholders. Consequently, on November 2, our Board of Directors temporarily suspended our return of capital program. As our cash flow increases, the indenture covenants will allow us to resume permitted additional distributions to shareholders. Now for a couple of comments on our revenue traffic and rep productivity growth rates. On a constant currency basis, our third-quarter 2015 revenue grew at an accelerated rate from the third quarter of 2014. It in fact grew by 12.1% and growth from the second quarter of 2015 to the third quarter of 2015 was sequentially 4.4%. Included in our third-quarter revenues and in our cost-of-network operations expense was a pass-through of $1.8 million in excise taxes, primarily payments to the Universal Service Fund. As more of our revenues become subject to these fees due to the open internet order and the clarification of the treatment of VPN services by the FCC, under U.S. GAAP, these amounts are recorded on a gross basis in our income statement. During the quarter, our network traffic grew at an accelerated rate of 11% sequentially from the second quarter of 2015 and by 39% on a year-over-year basis. Our sales force productivity per full-time equivalent rep increased from 5.6 units per month in the previous quarter to 6 units per rep per month in the third quarter. This again is substantially better than our long-term historical rep productivity numbers of 4.8 units per full-time equivalent per month. Our sales rep turnover rate was or 4.7% per month, a rate that was also significantly better than our long-term historical rate of rep turnover of 6.2%. Throughout this discussion, we will highlight several operational statistics, I will review in greater detail some operational highlights and trends and Tad's going to provide some additional details on our financial performance. Following our remarks we will open the floor up for questions. Now I'd like to turn the call over to Tad to read our Safe Harbor language.

Thank you, Dave. Good morning, everyone. This earnings conference call includes forward-looking statements. These forward-looking statements are based upon our current intent believe and expectations. These forward-looking statements and all other 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 forward-looking statements. If we use any non-GAAP financial measures during this call, you will find these reconciled for the GAAP measurement in our earnings release which is posted on our website at cogentco.com. Now I'll turn the call back over to Dave.

Thanks, Tad. Hopefully, you've had a chance to review our earnings press release. Our press release includes a number of historical metrics. Our EBITDA as adjusted increased by 9.9% from the second quarter to $34.4 million. Our EBITDA as adjusted margin increased from the second quarter by 170 basis points and was 33.4%. On a constant currency basis, our revenue increased by 12.1% from the third quarter of last year and sequentially increased by 4.4%. We continue to anticipate that our full-year revenue growth for 2015 over 2014 and for 2016 over 2015 on a constant currency basis will be within our guidance range of 10% to 20%. We anticipate that gains on the equipment transactions in 2015 will be approximately $5 million as compared to similar gains in 2014 of $11 million. Gains on equipment transactions were $1.2 million for the quarter as compared to $700,000 in the second quarter and $3.1 million in the third quarter of 2014. We expect that our EBITDA as adjusted margin expansion for 2015 versus 2014 will be below the 100 basis points that we had previously indicated due to continuing legal fees necessary for the support of net neutrality as well as reduced gains from equipment transactions. We anticipate incurring additional net neutrality costs for the remainder of 2015 and into early 2016. We expect our legal expenses related to net neutrality for 2015 will be less than that of 2014 and 2016 will be significantly less than 2015. But we're still unsure of the exact amounts and the timing of these payments. We anticipate that we will return to our historical rate of 200 basis points of EBITDA margin expansion on a year-over-year basis as adjusted for full year 2016 versus 2015. Our revenue and EBITDA guidance are intended to be long-term ranges and goals and are not intended to be used as specific quarterly guidance. Tad will now cover some additional details related to our results for the quarter.

Thank you, Dave and again, good morning to everyone. I'd also like to thank and congratulate the Cogent team for the results and their continued hard work and their efforts during another busy and productive quarter for the Company. On corporate and a net-centric revenue in customer connections. As a reminder we analyze our revenues based upon product class which is on-net, off-net and non-core and we also analyze our revenues based upon customer type. We classify all customers into two types, net-centric customers and corporate customers. Our net-centric customers buy large amounts of bandwidth from us in carrier neutral data centers and our corporate customers buy bandwidth from us in large multi-tenant office buildings. Revenue from our corporate customers grew by 6.2% from the second quarter to $60.6 million and grew by 18.3% from the third quarter of last year. Revenue from our net-centric customers increased by 1.7% to $42.4 million for the quarter and decreased by 4.7% from the third quarter of last year. The net-centric revenue decrease compared to last year was primarily due to the $4.3 million year-over-year negative impact of foreign exchange on our third-quarter 2015 revenues. Our European revenue is almost entirely net-centric revenue and accordingly is particularly vulnerable to the impact of variations in foreign exchange. On a constant currency basis, our net-centric revenue grew by 5% from the third quarter of last year as opposed to the decline. Revenue in customer connections by product class. Our on-net revenue was $75.1 million for the quarter which was a sequential increase of 4.3% and an increase of 5.7% from last year. Excise taxes included in our on-net revenues were approximately $1.3 million of the $1.8 million total for the quarter. And are on-net customer connections increased at an accelerated rate and increased by 3.2% sequentially and by 12.5% from the third quarter of last year. We ended the quarter with over 43,000 on-net customer connections on our network in our 2,221 total on-net multi-tenant office buildings and carrier neutral data centers. Our off-net revenue was $27.7 million for the quarter which was sequential increase of 4.4% and an increase of 13.8% from last year. Excise taxes included in our off-net revenues were about $500,000 of the $1.8 million for the quarter. Off-net customer connections also increased at an accelerated rate and increased sequentially by 4.8% and 21.1% from the third quarter of last year. We ended the quarter serving over 6,800 off-net customer connections and over 4,500 off-net buildings primarily in North America. Some comments related to pricing. Consistent with historical trends the average price per megabit of our installed base in our new customer contracts decreased for the quarter. The average price per megabit for our installed base declined, but at a slower rate and declined by 3.7% from $1.63 for the second quarter to $1.57 for this quarter. And the decline was 18.7% from $1.93 for the third quarter of last year. The average price per megabit for new contracts was $1 which was a 10.3% decline from the $1.12 for new customer contracts that were sold in the second quarter and about a 18% decline from $1.22 per megabit for new customer contracts sold in the third quarter of last year. These average year-over-year price declines, price per megabit declines, are also impacted by foreign exchange rates, as about half of our network traffic is in Europe. ARPU, average revenue per unit. Our on-net and off-net ARPUs actually increased sequentially slightly from the second quarter. Our on-net ARPU, including both in corporate and net-centric customers, was $586 for the quarter which was an increase of 1.1% from the $580 for the second quarter. Our off-net ARPU which was comprised predominantly of corporate customers, was $1,369 for the quarter which is a slight increase from the $1,365 for the second quarter. Turn rates, our turn rate for on-net customers was stable during the quarter and our off-net turn rate materially improved. Our on-net turn rate was identical for this quarter and was 0.9%, the same as the last quarter. And our off-net turn rate was 1.3% for the quarter which was a material improvement from 2.1% for the second quarter. Gross margin, our gross profit margin decreased by 90 basis points from the second quarter and our gross profit margin was 56.3% for the quarter compared to 57.2% for the second quarter and 57.9% for the third quarter of last year. Excise taxes, including the Universal Service Fund fees that were included in our cost of networks operations expense again were $1.8 million for the quarter and that has a negative impact of about 90 basis points of our gross margin for the quarter. Our EBITDA and EBITDA as adjusted. Our presentment of EBITDA excludes asset-related gains and includes all of our net neutrality fees. EBITDA and EBITDA as adjusted are both reconciled to our cash flow from operations in all of our press releases, including this one. EBITDA was $33.2 million for the quarter and our EBITDA margin was 32.2% which was an improvement sequentially of 130 basis points and a year-over-year improvement by 20 basis points. The impact of excise taxes was about 60 basis points on our EBITDA margin for the quarter, a negative impact. Our EBITDA as adjusted includes gains related to equipment transactions and EBITDA as adjusted was $34.4 million this quarter and the margin improved sequentially by 170 basis points to 33.4%. Asset-related gains and net neutrality fees had a material impact on our EBITDA and EBITDA as adjusted calculations and can also vary materially from quarter to quarter. Our legal and economic analysis fees associated with defending net neutrality were $800,000 for the quarter and our equipment gains were $1.2 million. We're also impacted by seasonal factors. Seasonal factors that typically impact our SG&A expenses and consequently our EBITDA include the resetting of payroll taxes in the U.S., the cost of our annual sales meeting, annual cost of living increases and the timing and level of our audit and tax services. These seasonal factors typically increase our SG&A expense in our first quarter. Our SG&A expenses for this quarter declined by $1.2 million from the second to third quarter, from a combination of lower bad debt expense from excellent customer collections and also from some of our cost control initiatives. EPS, our basic undiluted EPS was $0.07 for the quarter compared to $0.02 for the second quarter and $0 for the third quarter of last year. Some comments on foreign currency impact. The impact of foreign currency has reduced the proportion of our business reported in U.S. dollars and located outside of the United States to about 22%. About 17% of our third-quarter revenues were based in Europe and about 5% of our revenues are related to our Canadian, Mexican and Asian operations. Continued volatility in foreign exchange rates can materially impact our quarterly revenue and financial results. The average euro to U.S. dollar rate for the second and third quarter was flat at $1.11. The average euro to U.S. dollar rate for the third quarter of last year was substantially greater at $1.33. The foreign exchange impact on our revenue from the second to third quarter resulted in a $200,000 decrease to our revenue primarily due to the weakness in the Canadian dollar. From the third quarter of 2014 to the third quarter of 2015, the foreign exchange impact on our revenue was much more significant related to both the Canadian dollar and euro and was a negative $4.3 million. The average euro to U.S. dollar rate so for this quarter is approximately $1.13 and the average Canadian dollar exchange rate is $0.77. Should the average foreign exchange rates remain at current levels for this fourth quarter of this year, we estimate that there will not be material FX conversion impact from Q3 to Q4. The average euro to U.S. dollar rate for the fourth quarter of last year was $1.25, coming down from the $1.33 and the Canadian dollar was $0.88. Should those rates remain at current levels we estimate that the FX impact on a year-over-year quarterly basis will be a negative impact of about $2.7 million. Customer concentration, we believe that our revenue in customer base is not highly concentrated for the third quarter of 2015. No customer represented more than 2.3% of our revenues and our top 25 customers represent less than 8.2% of our third-quarter revenues. Capital expenditures, our CapEx decreased 37% on a sequential basis and by 56% on a year-over-year basis for the quarter. CapEx for the quarter was $6.8 million versus $10.9 million for the second quarter and $50.4 million for the third quarter of last year. Capital leases. Our capital lease principle payments are for long-term dark fiber IRU agreements only. These payments were $6 million for the quarter, $7.3 million for the second quarter and $7.3 million, the same amount for the third quarter of last year. Cash and operating cash flow, at the end of the quarter our cash and cash equivalents totaled $207.3 million and for the quarter our cash decreased by $17.2 million as we returned $34.6 million of capital to our stakeholders. During the quarter we paid $15.3 million for our third quarter dividend, $12.2 million was spent on stock buybacks and $7.1 million was spent on interest payments on our debt. Our cash flow from operations increased and was $23.4 million for the quarter, compared to $20 million for the second quarter and $16.1 million for the third quarter of last year. Capital leases, our capital leases IRU obligations again are for long-term dark fiber leases and typically have initial terms of 15 to 20 years or longer and often include multiple renewal options after the initial term. Our long- and short-term capital lease obligations totaled $130.9 million at quarter end. Debt ratios, our total debt, including our capital lease obligations was $596.4 million at quarter end and our net debt was $389.1 million. Our total gross debt to trailing last 12 months EBITDA as adjusted ratio was 4.57 at September 30 and our net debt to trailing last 12 months EBITDA as adjusted ratio was 2.98. Bad debt expense in days sales. Our bad debt expense for this quarter substantially improved attributing to the reduction in SG&A and was again less than 1% of our revenues and was only 0.6% of our revenue, an improvement from the 0.9% for the second quarter and 0.8% from the third quarter of last year. And finally, our day sales outstanding or DSO, for worldwide accounts receivable was again historically low levels from outstanding customer collections and was only 23 days at quarter end, improvement from the already low 25 days at the end of the second quarter. And again, our group of worldwide billing and collection team members have just done a fantastic job and I want to recognize and thank them. I will now turn the call back over to Dave.

Thanks, Tad. Now for a couple of comments about our network and its scale. The size and scale of our network continues to grow. We have over 823 million square feet of multi-tenant office space in North America directly on net. Our network consists of over 28,000 metro fiber miles and over 56,000 intercity route miles of fiber. The Cogent network is the most interconnected network in the world. We connect with over 5,510 networks, as of the end of the quarter, with slightly less than 35 of these being settlement free peers, the remainder being Cogent customers. We're currently utilizing approximately 30% of our lit capacity. We routinely augment capacity in parts of our network to maintain these low utilization rates. We added an additional data center and today have 51 data centers in our network, comprising 564,000 feet of rentable space. In summary, we believe that Cogent remains the low-cost provider of internet access transit services and VPN services to our valued customers and our value proposition remains unmatched in the industry. Our business remains completely focused on the internet and IP connectivity and data center colocation and provides a utility to our customers. We expect our annualized constant currency long-term revenue growth to be consistent with our annualized growth rates of 10% to 20% and our long-term EBITDA annual margin expansion rate to also be consistent with the 200 basis points that we've historically delivered of margin improvement. We have incurred material legal and economic analysis fees related to net neutrality and spent $800,000 in the third quarter of 2015 supporting the FCC in these issues. We believe and continue to believe that we will incur additional fees in 2015 and in early 2016, but these fees will be lower than 2014 and 2016 should result in even lower fees than spent in 2015. Our Board of Directors has approved yet another increase in our regular quarterly dividend to a quarterly rate of $0.35 per share. Our dividend increase demonstrates our continued optimism with regard to the cash flow capabilities of our business and our commitment to return capital to shareholders. We will be opportunistic in the timing and purchase of our common stock. At the end of the quarter, we had $47.8 million remaining under our current buyback authorization which is scheduled to run through the end of 2016. We however are currently limited in our ability to use that buyback due to certain debt covenants. We're committed to returning increasing amounts of capital, both through buybacks and through dividends to our shareholders on a regular basis. With that, I'd like to now open the floor for questions.

[Operator Instructions]. And our first question comes from Eric Pan of JPMorgan. Your line is open.

Just a little clarification on the U.S. excise tax, did it increase by $1.8 million sequentially? How should we think about that going forward? Is this one-time in nature or is this permanent? And is that pass through 100% of your customers?

So the answer is it did increase sequentially by about $1.7 million. It was a very small U.S.F impact previously. So in the final open internet order which was adopted on June 12, what the FCC did was defined the internet as meaning access to all points on the internet. The cost approximately 20% of our ports sold and about 12% of our revenues come from tunneled services, where Cogent is providing the VPN over the public internet to our customers, we felt that with this clarification, we needed to impose the U.S.F charge on our customers. The sequential increase was about $1.7 million. There was about $100,000 previously of certain excise taxes and it will be permanent. It will continue going forward and it is a direct pass-through and resulted in an increase in our cost of goods sold. And for that reason, it impacted our gross margin negatively by about 90 basis points and it impacted our EBITDA margin negatively by about 60 basis points.

Right and it looks like you will be able to achieve at least the low end of your revenue guidance after the strong Q3 results. What about your margin guidance for the year? Can you give us an update on that?

So on the revenue guidance, we will clearly be within the 10% to 20% year-over-year growth rate constant currency, even without the benefit of U.S.F in the latter two quarters. However, on the margin expansion, while we were anticipating about 100 basis points of EBITDA margin expansion this year, the reality is it'll probably be slightly down, primarily due to the arithmetic of the U.S.F hitting both our EBITDA margin and gross margin as well as a greater expenditure on legal than we had anticipated. We unfortunately had to prepare for litigation against two North American providers. One of them has recently entered a publicly announced agreement with us, the other is in final negotiations and we anticipate an agreement shortly. We continue to work with the three European providers who have also exhibited anti-competitive and anti-internet practices. We have seen changes in behavior from Orange and are in the process of trying to negotiate a permanent agreement. Telefonica and Deutsche Telekom have been a bit more reluctantly, although Telefonica has in negotiations and while the European Parliament passed net neutrality rules, that for the most part mirror those in the U.S., there are some subtle differences and we may be forced to litigate here in the U.S. against those providers. We hope not, but it's possible. So for those reasons, we believe that our margins will probably actually decline slightly on a year-over-year basis.

And last one from me, CapEx was down meaningfully this quarter. Was that due to the leverage restrictions? Or is this sort of the new level that we can expect going forward?

We have told investors repeatedly that we will bring our capital expenditures, that is both CapEx and principal payments on capital leases, back to the levels they were at in 2013 and then they would decline from that level. We did have a material spike-up in CapEx in 2014 for two very discreet reasons. One was a rapid expansion of our data center footprint. The second was the purchase of a completely parallel and redundant network in Spain that we ultimately rolled our traffic to and have exited the previous more expensive network. With those projects primarily behind us, you should expect to see our CapEx continue at a lower level. The covenants do not impact our ability to spend cash for CapEx as that remains in the bondholders collateral pool. The issue is we have a restricted payments basket of about $100 million that is currently trapped because our gross debt to EBITDA is above 4.25% or 4.25 to 1. Once we go below that, we will be able to access that builder basket and it will have continued to increase in size. We feel comfortable that we have adequate available capital in that basket to meet our dividend obligations, but have been maybe more cautious on our use of excess capital for buybacks. And for that reason, we did not want to be bound by the guaranteed commitment of $12 million a quarter.

And our next question comes from Scott Goldman of Jefferies. Your line is open.

Dave, I was just wondering maybe on the return to cash, it seems to have caught a number of people by surprise this morning and maybe if we all did our homework a little bit better that might not have been the case. But just wondering I mean it seems like this covenant's obviously been out there and last quarter you raised the target range for leverage. Is it, maybe just talk about how come this isn't something that maybe we hadn't seen coming and maybe could have messaged a little bit sort of going into this to reduce the surprise. And then secondly, maybe, you mentioned just a little bit on the interconnect side of the equation some of the agreements, but maybe you could talk a little bit about the contributions that you are seeing from the agreements that you have signed over the last few months. How they ramped up for the quarter, maybe what those signed agreements have contributed to the improvement in the overall growth and net-centric growth of the business? Thanks.

Sure, so let me start with our covenants. You know, those have been very clearly messaged. We've disclosed them publicly. I've commented them on those covenants repeatedly at conferences and in conversations with investors. Just to remind investors, we have two instruments outstanding, our senior secured notes and our senior unsecured notes. Those two instruments total $450 million. They have three significant covenants that are identical in both notes. They limit the Company's aggregate gross secured leverage to 3.5 to 1, gross leverage of 5 to 1 secured and unsecured and then finally, there was an initial builder basket and one that continued to increase in size and it was available to the Company once our gross leverage is below 4.25 to 1. If you remember, about a year ago, we went through a reorganization in order to have access to the builder basket that was going to become trapped once we issued our unsecured notes and we explained that clearly to investors the purpose of that reorganization was just to preserve the builder basket that was built but that the new builder basket was going to be unavailable to us. On a going-forward basis, we continue to delever. We have capital available that is available for investment in the Company. There are no restrictions whatsoever in investments that we make within the Company. However, for capital that is returned to our equity holders, that needs to come out of that restricted basket. And while we have surplus available in that basket, because of the factors that we've just talked about, more legal expense and also lower equipment sales, we felt that it was prudent that we temporarily suspend the buyback program. Our commitment remains to stay within the targeted leverage range of 2 1/2 to 3 1/2 times EBITDA. We're at 2.98 today. We expect that within several quarters, we will be able to once again access that $100 million of builder basket that exists today to distribute that to shareholders and that basket will actually grow in size as we continue to generate increasing amounts of free cash flow. So this is a temporary suspension in a commitment. The general intention to continue to return free cash and to remain within the guidance range is unchanged and I think these have been pretty clearly communicated to investors, so I apologize if you or anyone else felt or thought they were surprised. Now with regard to your interconnection question, we have signed multiple interconnection agreements. We continue to see improvement in port capacity. We remain completely congestion free connecting to large networks such as Comcast and AT&T. While we have an agreement in place and Verizon has been rapidly augmenting port capacity, there still does remain some congestion between our networks in certain markets. And there is another round of port augments in place. We have recently announced a binding multi-year agreement with Time Warner Cable. Those augments are underway. Those ports today remain totally congested. We've also seen companies such as Orange augment ports which have been helpful. As a result of the augments that up been put in place, you have seen our sequential growth rate in traffic go from 5% in Q2 versus Q1 to 11% and our year-over-year growth rate go from 33% to 39%. As a result of that, coupled with the price declines that Tad indicated, roughly 18% year-over-year which is slightly more moderate than our historical run rate, we have seen our net-centric revenue growth accelerate. So our net-centric year-over-year constant currency revenue growth rebounded from a trough of 3% in Q2 to about 5% in Q3. Now our sequential growth rate was about 1.7% on a constant currency basis. It is true that about 48% of our net-centric revenues and traffic come from outside of the U.S. We're a global provider and that does dampen the reported number, but we anticipate continued reacceleration in our traffic growth and quite honestly reacceleration of global traffic growth for the entire internet as the open internet order and the European Parliament's order take effect.

Thank you. Our next question comes from Walter Piecyk of BTIG. Your line is open.

Can you just kind of give us some color on Netflix in terms of how their growth is relative to the rest of the Company? I think there's been some, I don't about concern, but just when the whole agreements were set with Comcast, Verizon, whoever, that, how much of the traffic would come back? Can just kind of frame up what type of growth you're seeing from them relative to other players out there? And then also, are there other companies that you see as kind of the next Netflix? Not necessarily, I'm not asking for a competitor to Netflix. I don't care about that. But just people that are obviously going to be growing an over the top type service that would be utilizing you guys as opposed to other potential alternatives there, that could generate the type of revenue which I know is still low single digits as a percentage of the mix but to generate that type of revenue and I assume, above average growth for you going forward. Thank you.

Sure and just as a reminder, our net-centric revenue is virtually all on net and therefore carries also very high contribution margins incrementally. We win business with all of our net-centric customers for one reason. We provide a quality product at the lowest price in the market. Now to deliver that quality, we need to have adequate connectivity to all networks. Because that connectivity has artificially been constrained, we saw a deceleration in our traffic growth and revenue growth for our entire net-centric base and for large customers such as Netflix. Now, independent third parties concluded that over the last couple of years, global internet traffic growth decelerated by about 20%. I believe that deceleration is directly attributable to this anti-competitive behavior and port blocking that has now been declared illegal in the major developed countries. And as a result, we're starting to see those parties reluctantly open up port capacity. Now for Netflix or any other large customer, they will look at Cogent and asked two question. One, is the quality of there? Meaning can the bits get to the end-user? And then two, is the price better? And we continue to grow both by winning business from new competitors as well as from existing customers by having those customers use us to connect globally to the whole internet and potentially to some of those other service providers who have entered bilateral agreements and charge more for those agreements.

So, Dave, just let me interrupt. So the concern is out there that maybe the agreements with Comcast and Verizon offer lower prices than what you can offer and as a result, they would continue to send more and more traffic there. It sounded like that is an inaccurate conclusion, that you still offer more attractive economics for Netflix and whoever else to put their increment traffic. The only issue was the quality of what they were getting based on the blocking correct?

That is absolutely correct, Walt and it came out in the open internet order hearings and in the Comcast, Time Warner hearings that our rate of price decline per megabit was 8X faster than Comcast's rate of decline in price per megabit. And we continue to lead the market in price decline and I don't believe Comcast is particularly focused for example on competing with Cogent in this highly commoditized transit market where they are a less efficient operator.

So how does Netflix's growth compare to your other net-centric customers?

Well, the only thing I can comment is that our largest customer increased from 2.1% of aggregate revenues to 2.3% of revenues, so therefore they grew faster than the rest of the banks. And then in terms of other OTT customers, we obviously take every customer seriously. We compete for business and I think we're the provider of choice. I think what you're also seeing, at least from some of the public content delivering networks, you're starting to see customers build their own CDNs and choose to use Cogent and that's why our revenue and our traffic are accelerating relative to some other providers.

Our next question comes from Colby Synesael of Cowen and Company. Your line is open.

Two questions, I guess two topics. One is on the suspension of the capital returns. So you mentioned that in several quarters that will come back. Is it going to be fits and starts, so it comes back for a few quarters and then we max out again and then we have to suspend for a few quarters or would you anticipate that when it comes back, that that's on a permanent basis? Also just to get a little cute, when you say several quarters, is it possible that that could imply the back half of 2016? And also just lastly on this topic, I appreciate that the Board is trying to be prudent but your expectations for a slowdown in equipment sales and expectations for legal costs really haven't changed that much, maybe $1 million or $2 million one way or the other, so I'm trying to understand, you mentioned that as this explanation for why you guys have chosen to prudently suspended it. That doesn't really, to me, add up in terms of understanding the rationale so I was hoping you could explain that a little bit more? And then I actually did just have one other topic I wanted to discuss after this.

All right, let me take the buyback. So three points, the first one being we're committed to using the debt markets as long as our cost of debt capital remains at or below where it is today to accelerate returns to our equity holders, both through buybacks and dividends. We have bought back over 10 million shares and over 20% of our outstanding float since doing our first buybacks. When we entered into the debt agreements, both the secured and senior unsecured notes, we understood that there was this restricted payments basket that would allow us to have capital flow outside of the collateral pool of the bondholders and that is in the form of either dividends or buybacks or a combination thereof. We're committed to a methodical, regular and increasing dividend policy and I think 13 quarters has proven that. Secondly, we did buy back shares overall since implementing this program a little faster than we expected and therefore we got to the 2.98 leverage a little faster than we expected. EBITDA has grown a little slower, but you're correct, Colby, that's marginal, that's not necessarily the reason. We have $100 million of the roughly $200 million on our balance sheet today that is trapped because we cannot access the builder basket. When we get access to that basket which will be when our gross leverage falls below 4.25 to 1, we will then have that money free to use to buybacks and dividends. Because the buybacks are meant to be opportunistic, we felt that it was prudent to preserve extra capacity for the continued dividends and dividend increases and we will look for the buybacks. Most likely based on reasonable projections, sometime in midyear 2016, we'll have access to that basket. When we get access to that basket, we're committed to staying within the leverage targets that we have outlined and returning that capital. So again, I don't think there's really been a change, but there's an acknowledgment that $100 million of the $207 million on our balance sheet today is effectively trapped for the benefit of the bondholders by the covenants that we have disclosed and agreed to and will become untrapped as the Company continues to grow it's free cash flow.

And then just my second topic, you had a change in your head of sales, I think since the last time you reported. We're under the impression that there was a philosophical difference between where your previous head of sales wanted to take the Company and perhaps where you or the Board wanted to take the Company. I was wondering if you could talk about that a little bit more.

Sure, so as many of you know, about 2 1/4 years ago we brought in Ernie Ortega to head our sales and Ernie did a phenomenal job for the Company in implementing a number of programs, training, systems, tools and modifications. Those was paid great dividends on the corporate side. As we saw port congestion becoming relieved, we knew that we needed to reaccelerate our net-centric growth rate and we needed to make changes to the net-centric sales organization in a similar way that we had done on the corporate side. And there was some difference about the type of rep, the type of quota, the targeting and focus of those reps and I think Ernie, myself and the Board did not see eye-to-eye. Ernie left the Company and we promoted Jim Bubeck, who had actually held that role in an interim period prior to Ernie joining the Company. He has been with us for over 15 years, ran our largest single region, the central region and now runs global sales for us and we feel very comfortable that Jim understands what it's going to take to both continue to deliver excellent performance on the corporate side and get that net-centric business back up to where it potentially can be now that we're seeing ports become available.

So no concerns or risk of slowdown in momentum or change in strategic strategy in terms of where the Company is today.

No, not at all and if you saw even what some of the reps that Ernie brought in, being asked to leave the Company, we actually still had a net increase in the quarter in both number of reps and full-time equivalents and we saw our rep productivity again reaccelerate.

And our next question comes from Jonathan Atkin of RBC Capital. Your line is open.

Yes. Two quick questions, on the net neutrality and related legal expenses, as we head into next year, how much of that would be related to Europe versus the U.S.? And then kind of following up on the topic of sales, I think one of the relatively recent initiatives has been to ramp up on indirect sales agents and so forth. And can you give us an update on where that stands? Thanks.

Sure. So two things, first of all, three of the eight providers that had been experiencing port blockage with Cogent and other providers are European providers although we interconnect with them both in Europe and the U.S. Two, the European Parliament just last week past a final set of net neutrality rules that were a combination of rules initially proposed by Parliament, the European Council and the European Commission and those are now in effect. In Europe it is difficult for a U.S.-centric company to litigate against European providers. I am expecting that we would utilize the more well-developed and better rules that were designed here by the FCC and probably litigate where we get a more immediate result in U.S. courts under the title II and open internet order which these companies are all bound by since they operate under our section 214 license in the U.S. So I hope we don't have to litigate against anyone. If we do, it will most likely be in the U.S. Now with regard to our indirect sales efforts, they continue at Cogent. They have increased as a percentage of our business but remain in low single digits when compared to the aggregate sales force and we're continuing to add resources to that group, continuing to develop a channel in indirect program and we expect that to be additive to our aggregate growth rate. We're very cognizant of our cost of revenue acquisition and as such, we really need to balance both direct and indirect to maintain that very efficient cost of revenue acquisition.

And our next question comes from Frank Louthan of Raymond James. Your line is open.

A couple of questions, I'm sorry if I missed this, but can you let us know what is the current headcount for the net-centric sales force and what do you expect that to be by year-end? And then on the internet course, can you give us a characterization of how many new ports are open or from some of the carriers since this spring when the internet order was put forth and what your expectation is for year-end? To get some idea of what the potential increase in volume might be. And then lastly, as your rep productivity has increased quite substantially over the historic metrics, not quite the same uptick in the sales, hasn't tracked quite the same, has your definition of the units changed at all over that time period? Is there anything that would've, would it make that different because it doesn't seem to have the same relationship that it used to? Thank you.

Sure, Frank. So three different questions, let me take them. We did not give the breakdown of net-centric versus corporate reps. Historically, Cogent has been roughly 67% corporate, 33% net-centric. As port congestion remained an issue, that ratio had changed and we have fallen to only about 27% or 28% of our sales force being net-centric. We're continuing to hire, both corporate and net-centric reps, growing the entire sales force and we should expect that over the next year or so, we return back to that roughly 2/3, 1/3 mix between corporate and net-centric reps. And those reps, net-centric reps, are either hired from the outside, solely to focus on that market or it is a career path for existing corporate reps who have shown a desire to get the additional training necessary to be a net-centric rep and then will sign up for both the larger quota and a higher base salary and variable salary if that's the case. The second point is, if we look at port capacity being opened up. Over the past 12 months, we have seen about a 30% aggregate increase in our peering capacity. We need to see more than that, particularly as traffic grew 39%. However, we've gotten a disproportionate amount of that increase from those providers who were congested. Unfortunately, we're prohibited from giving you exact numbers. I feel very comfortable under the NDAs that we've executed, to be able to indicate whether or not the ports are congested. And the FCC is continuing to monitor this and I actually was with the chairman yesterday and he spoke again about how a hedge provider, meaning a content company, buys connectivity to the interconnection point, the end-user customer buys connectivity to the interconnection point and his commission will not allow the interconnection point to become a bottleneck. And then finally, on the unit productivity, our definitions have remained consistent. They have not changed. However, the mix does shift over time. You know, if we have put in a renewed emphasis in selling space, meaning rack and power in our data centers which are lower ARPU units, we have seen an increase in our VPN services, in which we sell AV Land which counts as a unit. So that is part of the reason why ARPUs have come down and why unit productivity may actually outstrip revenue growth. But overall, we feel very comfortable that the sales force is continuing to improve in its efficacy, both in units and also in dollars sold.

And one thing just to clarify on the covenant issue, before when you had guided to the Board taking the leverage range to 2 1/2 to 3 1/2 times, it's my understanding of what you've told us today and given us all the details on the covenant, that at the time and still currently, 3 1/2 times has and was within the covenant range before. It's just that the Board's decided that they don't want to get quite that close for whatever reason. But that was always kind of within the covenant range anyway. Nothing has really changed in that front.

Nothing has change at all, Frank. The only thing that has changed is again $100 million could potentially be trapped and not available for us to distribute to the equity until we get to 4.25 to 1 on a gross basis and for that reason, we wanted to be prudent and make sure that there was enough unrestricted capital that was available to be given to shareholders to meet our increasing dividend obligation and preserve the flexibility to opportunistically buy back shares but not to push up against the limit.

And our next question comes from Nick Del Deo of MoffettNathanson. Your line is open.

Actually, I had two quick ones. So first, on the net-centric side, can you talk about the pace at which you're seeing traffic that might have been lost during the interconnection to speech migrate back? And how long after getting the ports with the given peered cleaned up, do think it takes for the traffic to get back to a normalized level?

Sure, Nick. So in the process, both companies tried to estimate how much port capacity is needed and in what locations. And it is more of an art than a science because we cannot, neither side can ascertain more than 100%, so when the port is congested, there's no measurement technique to understand how backed up it is. Now, in some of these providers, we've gone through multiple rounds of upgrades so we put in the set of ports, we took our best guess and they immediately filled up and we came back and went back for another round. Now as you can suspect, Cogent's view of need for additional capacity is probably much larger than the other side and we had to compromise and agree with something that kind of is in the order of internet traffic growth or maybe slightly faster than the growth rate, but not really recognizing Cogent's growth rate. But there is then a mechanism to then immediately kick off a second, a third, you know, fourth round of augments as necessary. And then for each of our customers, they sense that additional capacity available to them and for our larger net-centric customers, our sales teams sit down with the customers, sometimes as frequently as weekly and analyze where their traffic is coming from by geography, where it is going to by geography and to what network it's going. And then those customers then modified their routing announcements to accommodate that, but we still see the ability to accelerate our year-over-year traffic growth. Remember, while 39% is great and it's better than the industry and it's better than we have been doing, it's not our historic growth rate and of the question that can up earlier, we do think that over-the-top video will probably drive even faster growth going forward. So we think there's a long way to go until we're in a position where we can say every port to every provider has adequate capacity.

And then can you give us a quick update on how your data center sales have been going? In particular the recently added ones. I know you've had some time to develop and start implementing a marketing plan?

Yes, we absolutely have. We've certified about 15% of our sales force as data center specialists. We've done additional training and they get additional commission once they have that certification. We've seen our data center utilization rates climb back to about 29%. They had troughed when we added that bulk of 125,000 feet at around 25% to 26% and that's across the entire footprint. We did add one additional data center about 15,000 feet in this last quarter. It was actually a facility in San Diego, our second San Diego facility, kind of North of the city. And we will continue, I think, to try to fill up the space, since it has such high incremental margins.

And out next question comes from Michael Rollins of Citi. Your line is open.

Maybe just taking a step back, one of the questions I think for the business model and you talked a little bit about what you're doing on sales and distribution earlier in the call, but how you were looking at the cost of acquisition. And to get that next 5%, 10% of penetration whether it's in your corporate segment or your net-centric segment, will it cost you more in acquisition cost relative to history to get those next dollars of revenue? And how we should think about the evolution of the cost structure in that context. Thanks.

Well quite honestly, that was probably the biggest philosophical difference between Ernie and myself and part of the reason why I think Jim has stepped into his new role and in fact, in the most recent quarter, our cost of revenue acquisition did decline slightly. It had been trending up under Ernie and we look at the efficacy both in terms of units per rep and dollars spent to acquire the revenue. That was part of the discussion around balancing channel and indirect versus direct and I feel that there is a natural tailwind in both parts of our business, corporate and net-centric. Our addressable market remains large and we should be able to actually drive our cost of revenue acquisition down because of this tailwind and we feel quite comfortable even though we're adding a lot of people, we're getting those people productive quicker in their career.

Our next question comes from Michael Bowen of Pacific Crest. Your line is open.

Sorry if I missed this a little bit, but can you maybe give us a little bit more detail with regard to the sales force organization? I know that last quarter I think you said that you were bifurcating the sales force into hunters and harvesters and some of the more tenured net-centric sales rep received approximately 500 accounts etcetera, etcetera. Can you shed some light on any changes or stay the course with regard to what the new head of sales is going to do with the program? And then maybe a little bit more detail with regard to some of the things that have worked and some of the things that have not worked for net-centric?

Sure. So within our corporate business, one of the initiatives Ernie put in place two years ago was a bifurcation of more tenured and less tenured reps, less tenure focusing on smaller accounts, more tenured on larger accounts. About eight, nine months ago, we implemented that same bifurcation of the net-centric side. Our premier team which focuses on the very largest accounts that buy transit, continue to have improved success. We're adding to both that team and to the general net-centric sales organization. So the largest accounts are defined and therefore there's no hunting associated with them and there are 500 of those kind of equally split between content companies and large access networks. But because there is a net-centric addressable market of closer to 10,000 or 12,000 potential accounts, both access and content, the majority of the net-centric force continues to be a hunter force, not a force that is managing these very large accounts. And it does seem to be working. Our net-centric growth rate obviously accelerated materially. Going from a year-over-year constant currency growth rate of 3% to 5% in one quarter to me is really testament to a very effective sales organization and I think Jim will continue down this strategy.

And then maybe as a quick follow-up on that last point, with the nice pickup in the constant currency growth on net-centric and again I apologize if you'd already mentioned a little bit of this, but do you think that during the quarter and also going forward is this going to be more a result of the interconnection agreements? Or can you help us, you know, on a scale of 1 to 10, how much is this going to be from the sales force restructuring versus the interconnection agreements in the ports opening up?

Sure. So, in the short-term, I think the interconnection will have the greater impact because it's a physical blockage that gets removed. On the sales force, there is time to hire, develop and train those individuals. But again, we feel comfortable that in a world where there is no intentional port congestion such as there has been, our net-centric business should be able to return to historic growth rates of about 10% year-over-year. So we clearly have a ways to go and it's going to take a combination of both ports opening up and more and better salespeople.

And our next question comes from Jennifer Fritzsche of Wells Fargo. Your line is open.

I just wanted to ask on the capital leases, Tad had mentioned there's still a 15 or you had signed agreements of 15 to 20 year terms. Can you tell us where you are in these agreements? Because I think there is this fear that those rates will be increased as those IRUs come due.

We have gone through our IRU base and actually recast some IRUs, lowering rates in some cases, extending term in some cases and in fact, our average remaining life across our IRU base still is up close to 20 years. So we today have I think over 1600 or 1700 unique IRUs in the table. Now I'd probably have to defer to Tad on that the exact number. I know they're from 210 different suppliers and there's roughly about 310 unique agreements governing so some are multiplier IRUs under one agreement and the average life of those is about 20 years still. So of all of the issues we worry about, that's probably not the one and we've also shown that over the past few years, we've actually found situations where we could find alternate IRUs under better economics and we've elected to voluntarily exit certain agreements, the Spanish network, the bay area one we did a few years ago. There was a few small gain actually this quarter of about $1.4 million that came from us exiting a IRU primarily in France and northern Spain I think that resulted in a gain. So we're pretty comfortable with our IRU durability.

And our next question comes from James Moorman of D.A. Davidson. Your line is open.

Just real quick, in terms of, back to the buybacks. So you've temporally suspended the $12 million, but can you still buyback if you want to if there's some, to be opportunistic? And have you thought more about going to more of an opportunistic rather than doing the mandatory every quarter, just more or less buying that buying back when you see it being, when your stock goes down? Thanks.

So first of all, yes. We do have an authorization of $47.8 million remaining, so we can go in and do buybacks. We do however have to be cognizant of how much capital we have available to return to shareholders outside of the $100 million that is trapped by the current restricted payments basket task. We have about $60 million available today, but we also know that it will probably be a quarter two or maybe more until we can get access to that additional $100 million. And remember that number builds as the leverage test continues to come down. But as a result of that, I think for the next couple of quarters, just to be clear, we're going to make sure we have enough dry powder to absolutely honor our growing dividend commitment and buybacks will be opportunistic. We can do some but for the next quarter or two, we're going to be a little more prudent than we've been and therefore we felt the absolute guarantee of $12 million no longer made sense.

And our next question comes from James Breen of William Blair. Your line is open.

This is Louis Depalma on behalf of James Breen. Dave, the corporate business continues to quietly do well. With your corporate customers increasing utilizing over-the-top services and accessing cloud applications, has there been higher demand for connections greater than your most popular 100 meg service?

Absolutely, so about two years ago, we introduced a corporate 1 gig product, we actually do have a corporate 10 gig product as well that sold on kind of an individual case basis. We have seen our average corporate users utilization rate go up to about 16% of the port capacity that they've bought and you're absolutely correct, enterprises and SMBs are increasingly putting applications remote in either public or private cloud environments. And that is a key driver of our corporate business and I think that tailwind has helped us deliver, the mid-teens type of year-over-year growth for the past couple of years. And as much as some of the access network operators tried to retard internet growth, they really did not have the ability to retard these corporate users. They were really focused more on the OTT for the residential users, but I think as the open internet order really takes hold, you'll actually see an acceleration of OTT, not only on the consumer side, but also on the enterprise side.

And our last question comes from Ana Goshko of Bank of America. Your line is open.

So Dave, wanted to know if you've considered amending the covenant in those bonds, either through an amendment process, I think typically it takes the majority or some kind of refinancing where you did a tender and consent or otherwise redeem those funds?

Yes, sure, Anna. So the answer is we have thought about it. All of those have a cost associated with them and at this point we concluded that we're better off using the restricted basket that is available to us and growing into the other basket rather than spending the money either through a refinancing or through an amendment and consent. While we've talked to a number of our bondholders on a regular basis and believe we could get those done, they would have a cost and at this time, we did not see the benefit in doing that. But going forward we'll preserve that as an option. But we have enough available cash, both from operations and in the basket to meet our dividend and do some modest buybacks and I think within a few quarters we'll be back into being able to access that additional $100 million. So I think this kind of middle of the road path makes a lot of sense.

Thank you. And at this time I would like to turn the call back over to management for closing remarks.

Well again, I apologize for a call a little longer than I think people like, but we try to answer all the questions thoroughly. I want to thank everyone and particularly, I want to thank the coach and team for a phenomenal job. I want to thank the regulators for understanding how important the internet is to society and helping us get to the point where we can see reaccelerated growth and again, I want to thank everyone for their support and take care and we'll talk soon.

Ladies and gentlemen, thank you for participating in today's conference. This does conclude the program. You may all disconnect. Everyone have a great day.